How To Do Demand Planning

Demand planning can at times seem like an arduous and overwhelming process. There are lots of moving parts in the supply chain that can shape customer demand—and even more that can shape a forecast. If those moving parts aren’t kept in check, you can easily find yourself being plagued by the pitfalls of inaccuracy, such as over-forecasting and under-forecasting demand. From there, it snowballs: over-forecasting and under-forecasting lead to excess inventory or stockouts, and therefore rush charges, spoilage, high storage costs, and last-minute transport fees. It’s obvious that your demand planning process isn’t something you can afford to get wrong.

At the end of the day, though, demand planning boils down to this: understanding the true demand for your product and aligning your supply with it. Having a clear understanding of high-level moving parts in the supply chain is what keeps you from leaving money on the table. And, ultimately, that understanding ensures that your supply chain is at peak performance so that the right amounts of your product are delivered at the right place and time.

So how do you create a demand plan that works for your brand?

Below, we demystify all of those moving parts, such as...

• The kind of supply chain data you need for the foundation of your demand planning
• Manual vs. automated planning with spreadsheets and software
• The importance of data science and how a DS team can change the game
• How to use forecasting methods in conjunction with actionable insights
• The direct value of good demand planning on your supply chain

...giving you all the tools you need to build your demand planning process.

data

It all starts with data

Data is the lifeblood of a demand plan and, more specifically, a demand forecast. Ultimately, it constitutes the building blocks that get you closer and closer to perfect accuracy. But in a sea of data, what do you need? From which sources, during which time periods, and of what quality?

In other words, what data will give you the most accurate forecasts?

The answer lies, first and foremost, in your historicals. Historical sales data is a rich resource for discerning patterns, seasonality, and more. By identifying and following those sales patterns, you can gauge what future sales will look like if said patterns continue. After all, you can only know where you’re going if you know where you’re coming from.

Historical data—and a lot of it—is important, yes, but another key characteristic of the data is cleanliness. An accurate demand plan hinges on data that’s more or less free of gaps, mismatches, incompletions, inconsistency, etc. As we like to say at Unioncrate: garbage in, garbage out. And the opposite of garbage? Data that’s well-organized, timely, precise, consistent, and complete. Precise data leads to precise forecasting.

manual-automated

Manual vs. automated planning

Once you have your data ready to go, it’s time for forecasting. Forecasting is the main output of the demand planning process, and once you get the cold hard numbers, you can use them to make smarter, more informed decisions.

How do you approach demand forecasting? There are a couple of different routes you could take to build out your demand forecasts, and each forecasting method has its pros and cons.

Spreadsheets/Business Intelligence Tool

With this route, manual work is the name of the game. In the past—and much of the time, still in the present—planners have had to manually build forecasts from the bottom up, with their own human intelligence and the sales team’s educated guesses as their guide. What’s great about this option is that, first and foremost, it’s cheap, and with enough time anyone can learn Excel. More notably, though, it’s easy to customize visualizations exactly the way you want them to look.

However, the spreadsheet method takes time—lots of it. Not only is building everything from scratch tedious, but so is doing a customer demand analysis of all the factors that can inform a forecast: sales orders, purchase orders, inventory levels, promotions, velocity, sales reports from retailers or distributors, and so much more. Plus, manual data crunching creates room for human error, which takes even more time to go back and fix.

Demand Planning Software

A demand planning software, on the other hand, computes data on a humanly impossible scale to create demand forecasts from the bottom up, automatically. The supply chain factors that can affect a forecast—sales orders, purchase orders, velocity, etc.—are suddenly analyzed with the click of a button, along with seasonality, consumer trends, and promotional impact. If the forecasting software employs artificial intelligence, it can quickly deploy demand sensing technology to get a sense of actual demand, as well as identifying patterns and learning from any changes in supply or consumer behavior. Other benefits include centralization, a 360-degree single source of truth that improves collaboration, reduction of back-and-forth communication between a demand manager and adjacent team members, and elimination of human error. All of this leads to high forecasting accuracy—which translates to money saved, hours of freed-up time, and an overall-simplified planning process.

There are cons to supply chain planning software, too, but they’re usually less impactful than the cons of spreadsheets. First, there’s the cost. Supply chain software can come at a price, though it often still costs less than hiring more people to your supply chain teams. In addition to that, the money supply chain software can save you in the long run with accurate forecasts leads to less wasted capital to begin with. Second, there are still lots of misconceptions surrounding demand planning software, including long onboarding times, difficult adoption, and lack of benefit for smaller companies. All of these misconceptions can make it hard to get the necessary stakeholders on board, but with the right software and the right training, any supply chain team can pick it up quickly.

‍

[Related: How To Choose Demand Planning Software]

‍

data-science

The key role of data science

Even though there are different options for forecasting customer demand and building a demand plan (and a supply plan), there’s of course a common denominator—no matter which route you choose, getting into the weeds with your data is an absolute necessity. And getting into the weeds involves data science.

With software, you can have a whole data science team at your disposal. What does a data science team do, exactly? They study datasets, models, and metrics in order to optimize forecast accuracy. The datasets can look like anything from historical sales, demand volatility, and intermittent demand to adjacent factors like panel data, point-of-sale data, and weather, the impact of which many brands don’t have the time or resources to consider. Data scientists take this data and look for signals, patterns, trends, and shifts. If you’re paying for a supply chain planning software that comes with a built-in data science team, what you’re ultimately paying for isn’t just accurate forecasts. It’s expertise—especially for smaller brands who don’t have the resources to bring on a whole internal team.

With manual supply chain planning or a BI tool, figuring all of this out can take time—a lot of it. A demand planner doing things the manual way may not know what to look for in the data, and they probably don’t have the extra time to learn statistical forecasting and the rest of the data science field. Plus, on top of taking hours, manual data-crunching creates a whole lot of room for error. At the end of the day, only trained data scientists can properly decipher your sales data.

insights

Now, take your forecasts and make a plan

The demand forecasting process is just a function of demand planning; the real meat of demand planning is taking an accurate forecast and deciding what to do with them. Will you transfer your inventory for peak inventory optimization? Will you reschedule a production run? What about jump-starting a new marketing promotion? Although a demand forecast points you in the right direction, there are still plenty of paths you could take your business down. And this is why effective demand planning is bigger than just accurate forecasting; it’s about using the forecast as a jumping-off point to take your business to new heights. From there, your forecasts can trickle down into your supply planning, and the whole business benefits.

A key aspect of planning apart from forecasts is actionable insights. This could look like factors that contributed to past sales forecasting errors, whether it’s supply constraints or shipping issues. Or, it could look like sales drivers, the factors that drive the sales themselves (instead of the errors) that include everything from inventory management and seasonality to distribution and obsolescence. You can make even better decisions based on these insights when they’re spliced and diced by brand, product, customer, channel, and ship-to.

Another important insight to ascertain is how much your actuals deviated from your forecasts. Being able to see concretely the quality of your performance, with a number and a percentage, enables you to move forward confidently and with more clarity. A forecasting model gets you there in terms of determining these concrete numbers, but it takes a human brain to go above and beyond the forecasts to decide what to do next. That’s demand planning.

value

The value that demand planning brings

All of this can be so tedious that sometimes, you forget why you’re creating demand plans in the first place. But accurate demand planning implementation is essential if you want your brand to grow fast and reach its full potential. Here are just some of the benefits of a great demand plan in action:

• More time to focus on sales and growth
• Faster and more informed decision making
• Improved service levels
• Increased negotiating power with suppliers
• More precise production and ordering
• Reduced manual work (and no more data crunching)

These benefits naturally translate into more money in your bank account, which could look like this (and so much more):

• Less money spent on resources
• Reduced over- and under-spending on raws and components
• Reduced working capital costs
• Maximized efficiency of line capacity
• Reduced logistics costs

When accuracy metrics translate into dollars, it’s proof that your demand planning is headed in the right direction.

Great demand planning = growth for your business

The foundation of your demand planning matters. It sets the tone for your business at large, and a more structured foundation—no matter the size of your business—makes way for faster, more efficient growth. Ultimately, an effective demand planning process requires you to simplify things: The key is taking a complicated, manual, and archaic process and adding in automation, visibility, and agility with the right planning process. Do this, and your operations will be golden.

‍

‍

About Unioncrate

Unioncrate is an AI-powered Integrated Business Planning (IBP) platform that delivers demand forecasts with unmatched accuracy, collaborative visibility, and actionable intelligence—enabling CPG brands to plan and execute agile supply chain strategies at the click of a button.

‍